insects Effects of Oxalic Acid on Apis mellifera (Hymenoptera: Apidae) Article Eva Rademacher *, Marika Harz and Saskia Schneider

Transcription

1 insects Article Effects Oxalic Acid on Apis mellifera (Hymenoptera: Apidae) Eva Rademacher *, Marika Harz and Saskia Schneider Institute Biology/Neurobiology, Freie Universität Berlin, Königin-Luise-Str , Berlin, Germany; (M.H.); (S.S.) * Correspondence: e.rademacher@fu-berlin.de; Tel.: Academic Editor: Brian T. Forschler Received: 12 June 2017; Accepted: 2 August 2017; Published: 7 August 2017 Abstract: Oxalic acid dihydrate is used to treat varroosis Apis mellifera. This study investigates lethal and sublethal effects oxalic acid dihydrate on individually treated honeybees kept in cages under laboratory conditions as well as distribution in colony. After oral application, bee mortality occurred at relatively low concentrations (No Observed Adverse Effect Level (NOAEL) 50 µg/bee; Lowest Observed Adverse Effect Level (LOAEL) 75 µg/bee) compd to dermal treatment (NOAEL µg/bee; LOAEL 250 µg/bee). The dosage used in regular treatment via dermal application (circa 175 µg/bee) is below LOAEL, referring to mortality derived in laboratory. However, treatment oxalic acid dihydrate caused sublethal effects: This could be demonstrated in an increased responsiveness to water, decreased longevity and a reduction in ph-values in digestive system and hemolymph. The shift towards stronger acidity after treatment confirms that damage to epilial tissue and organs is likely to be caused by hyperacidity. The distribution oxalic acid dihydrate in a colony was shown by macro-computed tomography; it was rapid and consistent. The increased density individual bee was continuous for at least 14 days after treatment indicating presence oxalic acid dihydrate in hive even long after a treatment. Keywords: Apis mellifera; Varroa destructor; oxalic acid dihydrate; toxicity; tolerance; sublethal effects 1. Introduction Oxalic acid dihydrate (OAD) is one most important organic acids used for control Varroa destructor. It has been known to be effective against parasite since end 20th century [1]. The European Group for Integrated Varroa Control developed OAD for final application stage in beekeeping [2,3]. Three different application methods OAD exist: trickling, spraying and evaporation. There principal points to be considered concerning medical treatment honey bee colonies: tolerability ingredient to bees, and toxicity to mites, as well as its distribution in colony, which again is directly affects toxicity and efficacy a substance. The trickling method OAD combines high efficacy against V. destructor and low bee mortality. The tolerability in bee colony has been documented in a concentration 3.5% (w/v) OAD and a dose ml per colony for Central Europe [4]. Approval as a veterinary drug has been given in many countries worldwide over recent years, for Germany in 2006 [5]. So far, toxicological data on individual bees in laboratory out combinatory effects has not been available. The mode action in colony was only partially clarified. In order to get a better understanding we tested toxicity OAD after dermal or oral application in laboratory. Our aim was to define no observed adverse effect level (NOAEL) and lowest observed adverse effect levels (LOAEL) for OAD, including a safety margin for dosage used in practical beekeeping. With this focus in mind, it is most important to look primarily at dosage range up to 10% bee Insects 2017, 8, 84; doi: /insects

2 Insects 2017, 8, mortality, as higher mortality rates not acceptable for practical beekeeping. Therefore, we decided not to test very high dosages as it is unreasonable to kill large numbers bees for non-relevant information. Furrmore, we wanted to understand which sublethal effects can be found and how OAD is distributed in colony. 2. Materials and Methods 2.1. Laboratory Tests: Treatment Individual Bees OAD Investigation Lethal Effects Acute Oral and Dermal Toxicity The toxicity tests were conducted during August and September using Apis mellifera carnica bees from our apiary at Institute Biology/Neurobiology, Berlin (Germany). The colonies were managed according to good beekeeping practice. Worker bees were recruited from brood combs by brushing about 100 individuals into small cages. The test bees were approximately five to ten days old. The cages made wood (105/65/120 mm length/width/height) bee wire and a glass plate on sides. The caged bees were kept under laboratory conditions in dark at 22 C and 65% relative humidity, draft-free. The bees formed small clusters and were starved for 24 h to ensure an even distribution food before being treated OAD. After application, y were held in small groups 10 bees per cage and received food (Apifonda, Südzucker AG, Mannheim, Germany) and water ad libitum. OAD (Caelo, Hilden, Germany) dissolved in sucrose solution (50% w/w) was applied to bees individually using two application forms: trickling 5 µl OAD solution onto abdomen (dermal) or feeding 10 µl OAD solution (oral). Topical treatment 5 µl was maximum volume solution applicable out considerable loss agent. In feeding trials, 10 µl solution allows feeding even high dosages by lower concentrations, so y were well accepted by bees. The test design for both treatments was identical. Each dosage was tested on 30 bees (three cages per dosage and ten bees in every cage). Each trial was replicated at least once leading to a minimum number 60 bees tested for each dosage and treatment method. The control groups (three cages ten bees per cage, one replicate) were treated in same way but received only sucrose solution (50% w/w). The acute dermal toxicity test was conducted different OAD concentrations: 3.5, 4.25, 5, 7.5 and 10% (w/v) and dosages: 175, 212.5, 250, 375 and 500 µg OAD/bee, respectively. The acute oral toxicity test consisted concentrations respectively dosages: 0.1, 0.5, 0.75, 0.8 and 1% (w/v) corresponding to 10, 50, 75, 80 and 100 µg OAD/bee. Bee mortality data for test and control groups were collected 24, 48 and 72 h after respective applications. Directly after treatment, bees were observed for four hours and in time intervals up to 72 h for signs behavioral changes Investigation Sublethal Effects Responsiveness to Water and Ascending Concentrations Sucrose Solution The proboscis extension response (PER) was used to test bees responsiveness to water and ascending concentrations sucrose solution (ACSS). Bees were recruited out colony as described for previous experiments and individually marked. Bees were treated dermally 5 µl 3.5% OAD in sucrose solution (50% w/w) trickled onto abdomen, controls received sucrose solution only (n = 60). The proportion animals releasing a PER was calculated for water and six concentrations sucrose solution (0.1%, 0.3%, 1%, 3%, 10% and 30%). Each bee was tested twice: prior to treatment and 24 h afterwards. Between tests bees were kept in cages and starved for three hours prior to each test to ensure equal motivation. Solutions were applied to antennae a three-minute inter-trial interval. To ensure equal motivation, bees were starved for 3 h prior to tests. Without anestization, every bee was cfully moved into a little plastic tube allowing only head antennae and proboscis to move freely. Bees were tested for water and n for sucrose responsiveness

3 Insects 2017, 8, by applying first a drop water followed by ascending sucrose solutions to antennae a 3-min inter-trial interval. The proportion animals releasing a PER was calculated. Longevity under Laboratory Conditions To obtain bees same age, a brood-frame was taken from a colony and put into an incubator at 34.5 C. The hatched bees were removed daily. The young bees were kept in small cages and provided food (Apifonda, Südzucker AG, Mannheim, Germany) and water ad libitum. Pretests showed a mortality rate up to 60% when bees younger than five days were treated OAD; for this experiment bees at age five days were chosen and treated dermally 5 µl 3.5% OAD in sucrose solution (50% w/w) trickled onto abdomen or sucrose solution (controls), respectively (n = 50). Dead bees were counted and removed from cages daily until last bee had died. The test was repeated four times, a total 200 bees per group were treated. ph Values Digestive System and Hemolymph Worker bees were brushed from honey combs and individually treated OAD. Dermal treatment was conducted an amount 5 µl 3.5% OAD in sucrose solution (50% w/w), dosage: 175 µg/bee, n = 120); oral treatment was performed 10 µl 0.35% OAD in sucrose solution (50% w/w, dosage: 35 µg/bee, n = 120). The test animals were kept in cages as described above. In intervals 24, 48 and 72 h post treatment bees were frozen and subsequently dissected according to standard methods [6], intestinal parts (crop, ventriculus and rectum) being removed and transferred onto micro-slides for ph measurement. An Inlab Surface Electrode, which enables ph measurement very small amounts liquid (minimum 5 µl), was cfully placed onto different parts digestive system. The electrode was connected to a FiveGo TM ph meter (limits error: ±0.01 ph). The sampling hemolymph was conducted by removal front and hind wings from bee s thorax. A slight pressure on thorax enabled extraction hemolymph from wing base. The droplet hemolymph was absorbed micro-capillaries and immediately transferred to micro-slides for ph measurement. Hemolymph samples (minimum five bees per sample) were pooled to gain at least 5 µl liquid Computer Tomography Honey Bee Colonies: Distribution OAD Internal structures a bee hive can be demonstrated by computed tomography [7]. Two honey bee nucleus colonies (A. m. carnica) were used for a distribution test a macro-computed tomography scanner (macroct). The colonies consisted approximately 4000 individuals and had already formed a winter cluster. The treatments were conducted in November out brood. OAD (3.5% w/v in sucrose solution 50% w/w) was applied in recommended dosage (according to package instructions for use-oxuvar ) by trickling onto bees in bee space. OAD was mixed water-soluble contrast agent Unilux (Iopamidol, 370 mg iodine/ml) in a dosage 185 µg/bee. The contrast agent Unilux showed no bee toxicity in a previous study [8]. For in-hive visualization OAD distribution a macroct scanner (Xvision, Toshiba) was used (Figure 1). With a helical scanner, a distance 250 mm was examined ensuring colony was captured completely. The CT images were reconstructed a slice thickness 2 mm (Table 1). For visualization via 2D images and data analysis we used stw efilmtm LiteTM (MergeTM Healthc 2008). The 2D images allowed measuring density individual bees in Hounsfield units (HU). This density is directly related to amount solution applied to surface bee s bodies. In a defined a 100 cm 2 in central a comb, as well as in boundary a bee cluster, density single bees (n = 144) was measured over three combs. In one colony, measurements were conducted before application (control) and 10 min respectively thirty minutes after applying OAD. The second colony measurements were conducted before application (control) and 3, 7 and 14 days respectively after applying OAD (n 211/group). Only bees placed parallel to macroct sectional plane were quantified. The total dosage radiation for scan entire

4 Helix 250 mm Peak X ray voltage 120 kv X ray tube current 80 ma Total scan time 75 s Insects 2017, 8, 84 Matrix 512x512 Scan field view large 4 17 Display field view 480 mm bee colony was mgy, spread Window over width helix distance mm (125 slices 2 mm thickness). Compd to dosage 500 mgy Window reported level by [9] for biological 300 effects in Drosophila melanogaster, this dosage can be considered harmless Total scan todosage bees mgy Figure 1. Bee colony placed in a macro-computed tomography scanner (macroct) Statistical Analysis Table 1. Technical data for A. mellifera colony scanning. The statistical analysis was conducted using SigmaStat 3.0 stw. Results were tested for Parameter Resolution in bee mortality rate using chi 2 -test on values at all-time intervals. The dose Slice thickness 2 mm response curves obtained SigmaPlot Pitch 3.0 stw were basis 2.5 for probit analysis and derivation lethal dosage (LD) values. Helix For comparison 250 ph mm values normally distributed data t-test was used, alternatively Peak X ray voltage in case non-normally distributed 120 kv data Mann Whitney rank sum test was used. The XPER ray tube rates current were compd between 80groups ma for water responsiveness Total scan time 75 s and in groups for sucrose solutions using chi²-test and McNemar s test, respectively. The Matrix 512x512 longevity data were analyzed Scan using fielda Kaplan Meier view survival analysis, large Gehan Breslow. The density values bees in small Display colony field units viewwere analyzed 480 mm t-test. Regarding all statistical tests a difference was considered Window to be width when p-value 1000 obtained was lower than Window level 300 Total scan dosage mgy 2.3. Statistical Analysis The statistical analysis was conducted using SigmaStat 3.0 stw. Results were tested for in bee mortality rate using chi 2 -test on values at all-time intervals. The dose response curves obtained SigmaPlot 3.0 stw were basis for probit analysis and derivation lethal dosage (LD) values. For comparison ph values normally distributed data t-test was used, alternatively in case non-normally distributed data Mann Whitney rank sum test was used. The PER rates were compd between groups for water responsiveness and in groups for sucrose solutions using chi 2 -test and McNemar s test, respectively. The longevity data were analyzed using a Kaplan Meier survival analysis, Gehan Breslow. The density values bees in small colony units were analyzed t-test. Regarding all statistical tests a difference was considered to be when p-value obtained was lower than 0.05.

5 Insects 2017, 8, Insects 2017, 8, Results 3.1. Laboratory Tests: Treatment Individual Bees OAD Investigation Lethal Effects Acute Effects Acute Oral and Dermal Toxicity After dermal application OAD toxicity increased slowly during observation time 24 to 72 h. After 72 h application 175 and µg/bee, respectively, showed no effect. After application 250 µg µg OAD bee bee mortality was was ly higher higher (chi (chi 2 -test, 2 -test, p = p 0.003) = 0.003) than than at 175 at 175 and and µg/bee. µg/bee. In In dosages dosages and and µg/bee µg/bee mortality mortality increased increased to >20% to >20% (Figure (Figure 2). The 2). The NOAEL NOAEL (72 h) (72 for h) dermal for dermal application application was 212.5, was 212.5, LOAEL 250 LOAEL µg OAD/bee. 250 µg The OAD/bee. extrapolated The LD extrapolated 10 (72 h) isld (72 µgh) OAD/bee. is µg The OAD/bee. LD 10 48The h after LD10 application 48 h after application reached a higher reached value a higher thanvalue after 72 than h: after µg/bee. h: The µg/bee. LD The h after LD10 application 24 h after application exceeded this exceeded value but this could value not but becould determined not be from determined data from (Table 2). data (Table 2). Figure Figure Bee Bee mortality mortality rates rates after after dermal dermal application application OAD OAD during during three three test test intervals. intervals. Mortality Mortality rates rates different different lower-case lower-case letters letters ly ly different different (chi (chi 2 -test, -test, p 0.05). 0.05). After oral application, bee mortality occurred at relatively low concentrations compd to Table 2. Toxicity parameters after dermal and oral application OAD. dermal treatment (Figure 3). 10 and 50 µg did not cause a mortality rate ly different from control group, while 75 Dermal µg resulted Application in ly (µg/bee) higher bee mortality Oral Application (chi 2 -test, (µg/bee) p = 0.027). A total 100 µg killed 55% 48treated h animals after 72 h72 h. The NOAEL 48(72 h h) for oral application 72 h was 50, LOAEL was 75 µg OAD/bee. The LD10 obtained by probit analysis was 60.3 µg/bee. The LD10 LD h after NOAEL application reached n.d. higher b values than after 72 h: 68.1 µg/bee. 75The LD10 24 h after 50 application was expected LOAEL a to exceed this value n.d. b but this could not 250 be determined from 80 data (Table 2). 75 chi 2 -test p = p < p = a These values represent lowest dosage difference in bee morality compd to controls; b Table 2. Toxicity parameters after dermal and oral application OAD. n.d. not defined, no occurred after 48 h; p-values refers to first increase in mortality (LOAEL). Dermal Application (µg/bee) Oral Application (µg/bee) 48 h 72 h 48 h 72 h After oral application, bee mortality occurred at relatively low concentrations compd to LD dermal treatment (Figure 3). 10 and 50 µg did not cause a mortality rate ly different from NOAEL n.d. b control group, while 75 µg resulted in ly higher bee mortality (chi LOAEL a test, p = 0.027). A total µg killed 55% treated n.d. animals chi b after 72 h. The NOAEL (72 h) for oral application was 50, 2 -test p = p < p = LOAEL was 75 µg OAD/bee. The LD a 10 obtained by probit analysis was 60.3 µg/bee. The LD h These values represent lowest dosage difference in bee morality compd to controls; after application reached higher values than after 72 h: 68.1 µg/bee. The LD b h after application n.d. not defined, no occurred after 48 h; p-values refers to first increase in was expected to exceed this value but this could not be determined from data (Table 2). mortality (LOAEL).

6 Insects 2017, 8, Insects 2017, 8, Insects 2017, 8, Figure 3. Bee mortality rates after oral application OAD during three test intervals. Mortality Figure 3. Bee mortality rates after oral application OAD during three test intervals. Mortality letters different (chi -test, 0.05). rates different lower-case letters ly different (chi 2 2 -test, p 0.05). During experiment, changes in behavior were observed only after oral application dosages During experiment, changes in behavior were observed only after oral application dosages also causing increased mortality in this period ( 75 µg/bee). The bees were less active, showed also causing increased mortality in this period ( 75 µg/bee). The bees were less active, showed minor movement and slowly formed bee clusters at top cage. Shortly after application minor movement and slowly formed bee clusters at top cage. Shortly after application y showed increased self-grooming. Some bees also extended proboscis. y showed increased self-grooming. Some bees also extended proboscis Investigation Sublethal Effects Investigation Sublethal Effects Responsiveness to Water and Ascending Concentrations Sucrose Solution in in test test, The PER on water increased after treatment in test (McNemar s test, p 0.001) and control 0.01, 4). 4). bees (McNemar s test, p 0.01, Figure 4). However, bees treated OAD showed higher (chi²-test, 2 response rates to water than controls (chi²-test, p 0.001). Figure PER (Proboscis Extension Response) to water prior and after treatment: both groups show increased responsiveness after treatment (McNemar s test, p and p = 0.004,). Different lower-case indicate between groups, before and after treatment (chi²- letters indicate between between groups, groups, before before and and after after treatment treatment (chi 2 (chi²- -test, ptest, 0.001) additionally; marked asterisks ) p 0.001) additionally; additionally; marked marked asterisks. asterisks.

7 Insects 2017, 8, Insects 2017, 8, Insects 2017, 8, The The sucrose responsiveness was also influenced by treatments, in in control group group responsiveness The sucrose decreased responsiveness ly was also at 3% influenced and 10% concentration by treatments, (McNemar s in control test, group p 0.025, 0.025, Figure responsiveness 5). 5). decreased ly at 3% and 10% concentration (McNemar s test, p 0.025, Figure 5). Figure Figure PER PER to ascending to ascending sucrose sucrose concentrations concentrations from control from control bees: response bees: response is ly is ly decreased decreased Figure 5. PER at 3% to and ascending 10% concentration sucrose concentrations (McNemar s from test, control p = bees: and response p = 0.018), is ly at 3% and 10% concentration (McNemar s test, p = and p = 0.018), decreased at 3% marked and 10% concentration marked asterisks. asterisks. (McNemar s test, p = and p = 0.018), marked asterisks. Bees treated OAD showed increased sucrose responsiveness, ly at 0.1% (McNemar s Bees Bees treated treated test, p = 0.024, OAD OAD Figure showed 6). increased sucrose responsiveness, ly at at 0.1% 0.1% (McNemar s test, test, p = 0.024, Figure 6). Figure 6. PER to ascending sucrose concentrations from OAD-treated bees: response after treatment Figure is Figure ly PER PER to to increased ascending ascending at sucrose 0.1% sucrose concentrations concentrations (McNemar s from from OAD-treated OAD-treated test, p = 0.024), bees: bees: response response after after treatment treatment is ly marked is ly increased asterisks. increased at at 0.1% 0.1% concentration concentration (McNemar s (McNemar s test, test, p = 0.024), 0.024), marked marked asterisks. asterisks.

8 Insects 2017, 8, Insects 2017, 8, Longevity under Laboratory Conditions Longevity under Laboratory Conditions The highest proportional bee mortality occurred directly after treatment (test group) and The highest proportional bee mortality occurred directly after treatment (test group) and between 20th and 21st day after hatching (control group). In test group bees lived at least for between 20th and 21st day after hatching (control group). In test group bees lived at least for two days, for a maximum 31 days and on average for 6.4 days. The bees in control group lived two days, for a maximum 31 days and on average for 6.4 days. The bees in control group lived for for an an average average days days (min. (min. two two days, days, max. max days). days). This This can can be be demonstrated demonstrated in in trend trend survival survival curves curves (Kaplan Meier (Kaplan Meier survival survival analysis, Gehan Breslow, p 0.001, 0.001, Figure Figure 7). 7). Figure 7. Survival curves under laboratory conditions: bees in control group survived ly Figure 7. Survival curves under laboratory conditions: bees in control group survived ly longer than bees treated OAD (Kaplan Meier survival analysis, Gehan Breslow, p 0.001). longer than bees treated OAD (Kaplan Meier survival analysis, Gehan Breslow, p 0.001). ph Values Digestive System and Hemolymph ph Values Digestive System and Hemolymph Crop: The ph values crops content after oral OAD application were subject to slight variation Crop: The (Figure ph 8). values The average crops ph value content (±SD) afteroral oralapplication OAD application OAD (24 were h: 4.62 subject ± 0.4; to48 slight h: variation 4.64 ± 0.36; (Figure 72 h: 8) The± average 0.43) was ph lower value during (±SD) 72 after h than oral application control. After OAD dermal (24 h: application, 4.62 ± 0.4; 48 h: 4.64 ph ± values 0.36; 72 were h: 4.44 at 24 ± h: 0.43) 4.82 ± was 0.37 lower and 48 during h: 5.02 ± hremaining than control. at control Afterlevel. dermal Only during application, last phinterval values did were atph 24 subside h: 4.82 to ± ± and (72 h: h) The ± ph 0.4values remaining obtained at by control control level. group Only during were 24 h: last 4.86 interval ± 0.46; did 48 h: 4.80 ph± subside 0.44; and to h: 4.99 ± 0.38 ± (72 h). However The ph values obtained were by not control group were (Mann Whitney 24 h: 4.86 ± rank 0.46; sum 48 h: test, 4.80 p ± 0.44; 0.05), and and 72 h: 4.99 standard ± deviation However means is were comparatively not high. (Mann Whitney rank sum test, p 0.05), and standard deviation means is comparatively Ventriculus: high. The average ph values liquid ventriculus contents were ly reduced between Ventriculus: 24 and The 48 average h after oral phapplication values liquid OAD: 6.50 ventriculus ± 0.24 (24 contents h) and were 6.38 ± ly 0.32 (48 h, Mann- reduced Whitney-rank-sum-test, p = 0.002), respectively (Figure 9). During last interval, ph value between 24 and 48 h after oral application OAD: 6.50 ± 0.24 (24 h) and 6.38 ± 0.32 (48 h, averaged 6.59 ± 0.17 (72 h). After dermal application OAD, ph value was on average 6.52 ± Mann-Whitney-rank-sum-test, p = 0.002), respectively (Figure 9). During last interval, ph 0.17 (24 h). Within following intervals ph reduced ly to 6.40 ± 0.21 (48 h) compd value averaged 6.59 ± 0.17 (72 h). After dermal application OAD, ph value was on average control group (t-test, p = 0.002) and 6.48 ± 0.14 after 72 h (t-test, p = 0.028). The ph values 6.52 ± 0.17 (24 h). Within following intervals ph reduced ly to 6.40 ± 0.21 (48 h) obtained by corresponding control groups remained at a constant level: 24 h: 6.55 ± 0.2; 48 h: 6.55 compd ± 0.21; 72 h: 6.55 ± control group (t-test, p = 0.002) and 6.48 ± 0.14 after 72 h (t-test, p = 0.028). The ph values obtained Rectum: The by ph corresponding value liquid control rectum groups contents remained averaged at a4.98 constant ± 0.27 level: 24 h after 24 h: oral 6.55 OAD ± 0.2; 48 h: application 6.55 ± 0.21; (Figure 72 h: 10) After ± h, ph was ly reduced to 4.77 ± 018 (t-test, p < 0.001); after 72 Rectum: h, ph The reached ph value 4.83 ± 0.16 and liquid remained rectumly contents averaged different to 4.98 ± control group h after (t-test, oral p OAD = application 0.003). After (Figure dermal 10). application After 48 h, OAD, ph was ph ly value averaged reduced 5.09 ± to (24 ± h) 018 and (t-test, 5.11 ± 0.18 p < 0.001); (48 after h). 72During h, phlast reached interval, 4.83 ± 0.16 ph was and 5.10 remained ± 0.17 ly and different different to to control control group group (t-test, p = value 0.003). (t-test, Afterp dermal = 0.016). application The ph values OAD, obtained ph by value corresponding averaged 5.09 control ± 0.16 groups (24 h) were: and h: 5.05 ± 0.18 (48± h). 0.31; During 48 h: 5.05 last ± 0.24; interval, 72 h: 4.99 ± ph was 5.10 ± 0.17 and ly different to control group

9 Insects 2017, 8, value (t-test, p = 0.016). The ph values obtained by corresponding control groups were: 24 h: 5.05 Insects ± 0.31; 2017, 48, h: ± 0.24; 72 h: 4.99 ± Insects Hemolymph: 2017, 8, 84 OAD, 24 h after oral application, caused a reduction in average 9 16 ph value Hemolymph: hemolymph OAD, 24 h 6.72 after ± oral 0.31 application, compdcaused to a control group reduction (6.98 ± in 0.17; Mann Whitney average ph rank value sum Hemolymph: test, p hemolymph = 0.011, OAD, Figure 24 h after 11) oral ± During 0.31 application, compd remaining caused to control a intervals, group reduction (6.98 ph ± increased 0.17; in Mann Whitney average to 6.83 ph ± 0.22 value rank sum test, hemolymph p = 0.011, Figure ). ± 0.31 During compd remaining to control intervals, group (6.98 ph increased ± 0.17; Mann Whitney to 6.83 ± 0.22 (48 h) and 6.97 ± 0.27 (72 h). After dermal application, ph values measured at first and second rank (48 h) sum and test, 6.97 p ± = , (72 Figure h). After 11). dermal During application, remaining ph intervals, values measured ph increased at first to and 6.83 second ± 0.22 intervals were 7.00 ± 0.12 (24 h) and (48 h), respectively, and thus similar to control (48 intervals h) and were 6.97 ± ± ( h). (24 After h) and dermal 7.00 application, ± 0.08 (48 h), respectively, ph values measured and thus at similar first to and second control groups: intervals 6.98 ± 0.17 (24 h); (48 h). After 72 h, average ph value was 6.88 ± 0.15, groups: 6.98 were ± ± ( h); ( h) and ± (48 ± 0.08 h). After (48 h), 72 respectively, h, average and ph thus value similar was to 6.88 control ± 0.15, ly groups: ly 6.98 reduced reduced ± 0.17 in(24 comparison comparison h); 6.96 ± 0.26 to to (48 control control h). After group group 72 h, ± 0.21 average 0.21 (t-test, (t-test, ph p = value p 0.006). = 0.006). was 6.88 ± 0.15, ly reduced in comparison to control group 7.17 ± 0.21 (t-test, p = 0.006). Figure 8. Average ph liquid contents honeybee crop after oral and dermal treatment (n Figure 8. Average ph liquid contents honeybee crop after oral and dermal treatment Figure 52/group). 8. Average ph liquid contents honeybee crop after oral and dermal treatment (n (n 52/group). 52/group). Figure 9. Average ph ventriculus contents after oral and dermal treatment (n 68/group). Figure Significant 9. Average ph indicated ventriculus by asterisks contents (48 after h: Mann Whitney oral and dermal rank treatment sum test, (n p = 0.002; 68/group). 72 h: Figure 9. Average ph ventriculus contents after oral and dermal treatment (n 68/group). Significant t-test, p < 0.05); indicated by asterisks marked (48 h: Mann Whitney asterisks. rank sum test, p = 0.002; 72 h: Significant indicated by asterisks (48 h: Mann Whitney rank sum test, p = 0.002; 72 h: t-test, p < 0.05); marked asterisks. t-test, p < 0.05); marked asterisks.

10 Insects 2017, 8, Insects 2017, 8, Insects 2017, 8, Figure Figure 10. Average ph rectum contents after oral and dermal treatment (n 68/group). Figure Average Average ph ph rectum rectum contents contents after after oral oral and and dermal dermal treatment treatment (n (n 68/group). 68/group). Significant Significant indicated indicated by by asterisks asterisks (48 (48 h: h: t-test, t-test, 0.001; 0.001; h: h: t-test, t-test, 0.05); 0.05); Significant marked marked indicated asterisks. asterisks. by asterisks (48 h: t-test, p = 0.001; 72 h: t-test, p < 0.05); marked asterisks. Figure 11. Average ph hemolymph after oral and dermal treatment (n 26/group). Significant Figure 11. Average ph hemolymph after oral and dermal treatment (n 26/group). Significant Figure 11. Average indicated ph by asterisks hemolymph (24 after h: Mann Whitney oral and dermal rank treatment sum test, (n p = 26/group) ; 72 h: Significant t-test, p = 0.006); indicated by asterisks marked (24 h: Mann Whitney asterisks. rank sum test, p = 0.011; 72 h: t-test, p = indicated by asterisks (24 h: Mann Whitney rank sum test, p = 0.011; 72 h: t-test, 0.006); marked asterisks. p = 0.006); marked asterisks Computer Tomography Honey Bee Colonies: Distribution OAD 3.2. Computer Tomography Honey Bee Colonies: Distribution OAD 3.2. Computer The distribution Tomography OAD Honey in Beecolony Colonies: after Distribution topical application OAD was demonstrated by macroct scanning The distribution (Figure 12). The OAD control in measurements colony after achieved topical application a mean density was demonstrated value by ± macroct The distribution OAD in colony after topical application was demonstrated by macroct 93.3 HU. scanning The relatively (Figure high 12). standard The control deviation measurements was caused achieved by heterogeneity a mean density value bees body mass After HU. scanning (Figure 12). The control measurements achieved a mean density value ± 93.3 HU. The application relatively high OAD standard (10 min) deviation density was value caused increased by to heterogeneity a mean value bees ± body mass. HU, which After The relatively high standard deviation was caused by heterogeneity bees body mass. After application was ly OAD different (10 min) from density controls value (t-test, increased p to 0.001). a mean Thirty value minutes after application HU, which application OAD (10 min) density value increased to a mean value ± HU, which was mean ly value was different ± 89.5 from HU, ly controls (t-test, different p 0.001). compd Thirty to minutes controls after application and values mean achieved value after was min (t-test, ± 89.5 p HU, 0.001, ly Figure 13). different compd to controls and values achieved after 10 min (t-test, p 0.001, Figure 13).

11 The mean value ( HU) in central a combs 10 min after treatment The mean value ( ± 87 HU) in central a combs 10 min after treatment bees was comparable to bees in boundary a ( HU). However, mean bees was comparable to bees in boundary a ( ± 86.7 HU). However, mean density values thirty min after treatment were ly different compd to central a density values thirty min after treatment were ly different compd to central a ( ) and boundary a ( ) combs (t-test, 0.015, Figure 14). ( ± 77.8) and boundary a ( ± 97.1) combs (t-test, p = 0.015, Figure 14). InsectsIn 2017, 8, colony 84 examined for long-term distribution, control measurements achieved In colony examined for long-term distribution, control measurements achieved a mean density value HU. Three days after application OAD density value mean density value ± HU. Three days after application OAD density value increased to mean value HU and was ly different from control was increased measurement ly to a mean (t-test, different value 0.001, from Figure controls ± ). Seven (t-test, HU and and p was ). ly days after Thirty application minutes different after from mean application control value was measurement mean value was (t-test, HU p 0.001, and ± 89.5Figure HU, ly 15). Seven and HU, respectively, different 14 days compd after application ly to different controls mean compd and value values was to controls achieved± and after HU values min and (t-test, achieved p 0.001, ± after three Figure HU, days 13). respectively, ly different compd to (t-test, 0.001, Figure 15). controls and values achieved after three days (t-test, p 0.001, Figure 15). Figure 12. MacroCT scanner image: bees on comb, 30 min after application OAD and Unilux. Figure 12. MacroCT scanner image: bees bees on on comb, min min after application OAD and and Unilux. Visualization comb bees sitting on it: it: dark as show low densities and bright as show Visualization a comb bees sitting on it: dark as show low densities and bright as show high high densities. high densities. densities. Figure 13. Density values before (control) and 10 and 30 min after treatment (n 144/group; t-test, p Figure 13. Density values before (control) and 10 and 30 min after treatment (n 144/group; t-test, p Figure 0.001), 13. Density values before (control) marked and 10asterisks. and 30 min after treatment (n 144/group; t-test, p 0.001), 0.001), marked marked asterisks. asterisks. The mean value ( ± 87 HU) in central a combs 10 min after treatment bees was comparable to bees in boundary a ( ± 86.7 HU). However, mean density values thirty min after treatment were ly different compd to central a ( ± 77.8) and boundary a ( ± 97.1) combs (t-test, p = 0.015, Figure 14).

12 Insects 2017, 8, Insects 2017, 8, Insects 2017, 8, Figure 14. Density values comb as center and boundary: between Figure 14. Density values comb as center and boundary: between as occur as occur 30 min 30 after min treatment after treatment (n 72/group; (n 72/group; t-test, t-test, p = 0.015), p = 0.015), marked marked asterisks. asterisks. In colony examined for long-term distribution, control measurements achieved a mean density value ± HU. Three days after application OAD density value increased to a mean value ± HU and was ly different from control measurement Figure 14. (t-test, Density pvalues 0.001, Figure comb 15). as Seven center and and 14boundary: days after application mean between value was as ± occur HU min and after treatment ± (n /group; HU, respectively, t-test, p = 0.015), ly different compd marked to controls and asterisks. values achieved after three days (t-test, p 0.001, Figure 15). Figure 15. Density values before (control) and 3, 7 and 14 days after treatment: different lower-case letters indicate (n 211/group; t-test, p 0.001). 4. Discussion We found unequal results OAD toxicity after dermal and oral application ingestion OAD is much more toxic to bees than application on cuticula, which was relatively well tolerated. This corresponds our first report bee tolerability concerning OAD as a single compound [10]. Investigations toxicity ten conducted combined substances. This makes it difficult Figure 15. Density values before (control) and 3, and 14 days after treatment: different lower-case to comp Figure 15. se Density results values before our findings. (control) and Concerning 3, 7 and 14 days dermal after treatment: application different lower-case toxic effects different letters letters dosages indicate indicate reported in (n (n literature. 211/group; 211/group; t-test, Dosages t-test, p p < ) ). µg/bee did not cause mortality 4. Discussion after 48 h, this is less than half dosage µg/bee no observed mortality in our test. The described LD10 (48 h) µg/bee was derived from a combinatory effect OAD and We found unequal results OAD toxicity after dermal and oral application ingestion OAD is much more toxic to bees than application on cuticula, which was relatively well tolerated. This corresponds our first report bee tolerability concerning OAD as a single compound [10].

13 Insects 2017, 8, Discussion We found unequal results OAD toxicity after dermal and oral application ingestion OAD is much more toxic to bees than application on cuticula, which was relatively well tolerated. This corresponds our first report bee tolerability concerning OAD as a single compound [10]. Investigations toxicity ten conducted combined substances. This makes it difficult to comp se results our findings. Concerning dermal application toxic effects different dosages reported in literature. Dosages <100 µg/bee did not cause mortality after 48 h, this is less than half dosage µg/bee no observed mortality in our test. The described LD 10 (48 h) µg/bee was derived from a combinatory effect OAD and acetone on anaestized worker bees (CO 2 ) [11]. It was lower than LD 10 (48 h) OAD as a single agent in our trial, which reached a level µg/bee. A combinatory effect OAD and acetone on anaestized worker bees (CO 2 ) may explain explicitly higher responsiveness tested animals in cited trials compd to our findings. Or tests a polyhybrid subspecies A. mellifera resulted in a level out observed effect 400 µg/bee and a lowest observed effect level 600 µg/bee after 48 h [12]. In our experiments 175 µg/bee, corresponding to 3.5% solution (30 50 ml depending on colony size) used in beekeeping practice, did not cause mortality for individual bees, different from controls 72 h after dermal application. Oral application resulted in high bee mortality at relatively low concentrations compd to dermal treatment. Bees reacted much more sensitively to oral application OAD. The LOAEL for dermal application is higher than for oral application by a factor >3. To assess risk an applied substance, not only mortality but also effects on physiological processes and behavior after pesticide exposition must be considered as described [13]. Physiological effects concerning acetylcholinesterase and glutathione S-transferase activities, as described for exposure to fluvalinate, not documented after OAD treatment (3% in 32% sucrose solution w/w, 50 ml/colony) on pupae, newly emerged, nursery and forager bees [14,15]. Honey bee larvae, treated OAD by spraying (about 121 µg solution/larvae) undergo histologic changes accidental cell death leading to necrosis [16]. These effects may be causally responsible for brood death after application OAD (3% in 50% sucrose solution, two treatments during summer) in breeding colonies [17]. In beekeeping practice, acute damages to brood can be excluded when honeybee colonies treated OAD during brood free period. However, long term effects possible; OAD has been found in bee colonies even six months after topical application following regular treatment [18]. Long-term effects (up to four months after application) like a reduced amount brood in treated colonies (3% OAD w/v, 4 ml per comb side, spraying) have been reported [19]. After dermal and oral application, respectively, high dosages up to 1320 µg/bee, OAD was recovered in internal organs and hemolymph adult bees [20,21]. It is assumed that after dermal application cuticula can be penetrated by acid [21]. Also, pathological repercussions e.g., degeneration rectal epilium, malphigian tubules and ventriculus, have been described after dermal application [20]. C 14 -marked OAD could be found in abdominal structures worker bees after topical application into colony [22]. The treatment OAD caused sublethal effects on A. mellifera. First indications changes in ph internal organs and hemolymph after a single dermal application OAD a dosage 175 µg/bee have been provided [23]. The individual treatment honeybees OAD in our experiments changed ph-value intestinal parts and hemolymph. The ph values crop contents reflect acid application, specifically after oral ingestion. Significant could be found in ph values ventriculus 48 h after oral and 72 h after dermal treatment: These deferred reductions in tested time intervals could be caused by slower penetration acid through cuticula compd to direct oral intake. Differences in ph structure rectum and hemolymph were also found, ph reduction occurs faster after oral application. The pro a ph reduction in digestive system even after 72 h indicates a long disposition OAD

14 Insects 2017, 8, in bee. The returned balance ph measured in hemolymph 48 h after oral application also indicates a possible buffer capacity hemolymph. We could demonstrate that OAD moving through digestive system or penetrating cuticula modified ph structure honeybee s intestinal parts and hemolymph. The shift towards stronger acidity after OAD treatment supports that damage to epilial tissue and organs [20] may be caused by hyperacidity. It also corresponds to timescale recovered OAD after oral application [21]. The increased acidity can cause chemical burns which eventually lead to necrosis [16], however a threshold cannot be derived from our tests due to low dose applied. Furr sublethal effects could be demonstrated in a decreased longevity under laboratory conditions and increased responsiveness to water. Bees treated OAD died much sooner than bees treated sucrose solution. The increased responsiveness to water 24 h after OAD treatment indicated an acidosis bees, which y may compensate an increased uptake water. This assumption is supported by shift towards stronger acidity, found after OAD treatment conducted in measurements ph-values. These results indicate a general impairment bees after treatment. The treatment in autumn/winter affected primarily long-living winter bees which essential for winter survival and successful colony development in spring. Treatment during summer brood can lead to substantial brood damage [15] as described above. Even by treating artificial swarms or nucleus colonies, it cannot be certain that damages will not occur due to long-term exposure to OAD in colony [18,19]. It has been described that after dermal administration bees carry white deposits primarily on body [11,20] and later also on third pair legs [24]. Concerning sublethal effects in colony after individual dermal application 175 µg/bee, changes in behavior were found [24]: bees were less active, showed reduced brood c, increased grooming behavior, and y also had a reduced life span. In beekeeping practice OAD is mostly applied topically. The distribution in colony may occur on two pathways: (1) through oral intake and distribution via trophallaxis; and/or (2) contact between bees and also contaminated hive material. OAD vapors in colony seem to be minor importance as OAD has a low volatility [25]. However, colony treatment using 3.5% OAD solution (30 50 ml per colony, depending on colony strength, applied by trickling) corresponds to an individual dosage approximately 175 µg OAD/bee. This dosage is well tolerated by individual bees in laboratory and is well below lowest observed adverse effect level concerning mortality after dermal application in our test. OAD applied orally in laboratory causes high bee mortality. As mortality observed in treated colonies was not highly increased, OAD was probably ingested in relatively low amounts by grooming and/or trophallaxis. The oral intake OAD after treatment at colony level seems to be little importance. The threshold value derived in laboratory was refore not exceeded for most target animals. This corresponds findings that when OAD is applied in a solution high sugar content bee mortality increases [26]. This suggests that high sugar content leads to more ingestion. In laboratory trials, ly higher bee mortality was found when sugar was added to OAD solution [27]. In practical beekeeping, appropriate use OAD (on average 175 µg/bee) by topical application in field is relatively safe for A. mellifera on colony level, even when some individuals die. Based on our findings in laboratory, threshold first adverse effects (LOAEL) could be reached in colony when an overdose 43% for individual bee is applied. However, due to attractiveness sucrose solutions to bees y ingest solution even if it contains toxic substances. A sugar substitute e.g., glycerol, a high viscosity and desirable distribution can prohibit oral uptake by bees [9]. It could optimize application OAD, but so far it has not been approved as a veterinary drug ingredient for bees in combination OAD. High acaricidal efficacy after topical application was only found when bee to bee contact took place [11]. In furr trials, when body contact between bees was prevented, but trophallaxis

15 Insects 2017, 8, allowed, it has been shown that trophallactic interaction did not lead to mite mortality; bee to bee contact seemed to be primary route distribution OAD in acaricidal relevant dosages [28]. Due to only minor oral intake, systemic efficacy against V. destructor seems to be improbable. A systemic effect requires a transfer acid into hemolymph bee. OAD was found in hemolymph only after application very high dosages [20,21]. After application 3.5% OAD solution (175 µg/bee), as used in beekeeping, changes in ph structure were found, but OAD was not recovered in hemolymph single bees at a detection limit 2.5 µg [24]. This amount used in beekeeping practice seems to be too low to reach a systemic toxicity to V. destructor through ingestion. Therefore, we conclude that mode action in colony must be contact poisoning against V. destructor. In order to reach high efficacy, ingredient acting by contact must be distributed in colony. The distribution OAD was shown by macroct. The results roentgenoscopy showed high density values for individual bees in test, much higher than in control measurement. A good distribution was already achieved after 10 min; this could be documented in central and boundary as combs. Lower density values in central comb as compd to boundary regions obtained after thirty minutes reflected movement bees. After thirty minutes density was generally lower, which led us to assumption that OAD was now also spread to material, e.g., wall hive. Bees have constant contact hive material; refore, OAD can be distributed again onto bees, maintaining a long-term contact acid. OAD on hive material can be found even several months after application [18]. The macroct analysis demonstrated a rapid and consistent distribution OAD involving a reduction individual dosage over time. Even after 14 days, density bees was still ly higher than prior to treatment, indicating a potential efficacy at least up to 14 days. The results from field trials, where maximum efficacy against mites was reached ten days after treatment, support this assumption [29]. 5. Conclusions OAD used to treat varroosis A. mellifera shows a rapid and consistent distribution in colony for at least up to 14 days, and high efficacy against mite, but also lethal and sublethal effects. In practical beekeeping, appropriate use OAD (one topical application, on average 175 µg/bee) is relatively safe for A. mellifera at colony level, even when some individuals lost. However, ingestion leads to high mortality. The reported sublethal effects highly decreased longevity, a reduction in ph-values in digestive system and hemolymph, and an increased responsiveness to water. The shift towards stronger acidity after treatment confirms that damage to epilial tissue and organs is likely to be caused by hyperacidity. Pathological repercussions e.g., degeneration rectal epilium, malpighian tubules and ventriculus may also occur. These results indicate a general impairment bees after treatment. The treatment in autumn or winter affects primarily long-living winter bees which essential for winter survival and successful colony development in spring. Treatment during summer brood can cause substantial brood damage. Even when treating artificial swarms or nucleus colonies it cannot be certain that damages will not occur due to extensive exposure to OAD in colony. Long-term effects such as reduced amount brood in treated colonies have been reported. OAD is one most important organic acids used for control V. destructor. It is indispensable but must be dosed precisely and applied as seldom as possible to prevent sublethal damages which eventually lead to loss bees. Long disposition in bee hive can cause accumulation acid and refore induce furr damage. Acknowledgments: ir cooperation. We would like to thank Tierärztliche Klinik für Pferde Seeburg, Germany for Author Contributions: Eva Rademacher, Marika Harz and Saskia Schneider conceived and designed experiments; Marika Harz and Saskia Schneider performed experiments; Eva Rademacher, Marika Harz and Saskia Schneider analyzed data; Eva Rademacher, Marika Harz and Saskia Schneider wrote paper.

16 Insects 2017, 8, Conflicts Interest: The authors decl no conflict interest. References 1. Popov, E.T.; Melnik, V.N.; Matchinev, A.N. Application oxalic acid in varroatosis. In XXXII International Congress Apimondia; Apimondia Publication: Buchst, Romania, 1989; p Nanetti, A.; Büchler, R.; Charriere, J.D.; Fries, I.; Helland, S.; Imdorf, A.; Korpela, S.; Kristiansen, P. Oxalic acid treatments for varroa control (review). Apiacta 2003, 38, Rademacher, E.; Imdorf, A. Legalization use oxalic acid in Varroa control. Bee World 2004, 85, [CrossRef] 4. Rademacher, E.; Harz, M. Oxalic acid for control Varroosis in honey bee colonies A review. Apidologie 2006, 37, [CrossRef] 5. Rademacher, E. Oxalsäure als Tierarzneimittel zur Bekämpfung der Varroose legal einsetzbar. ADIZ/die biene/imkerfreund 2006, 2, 9. (In German) 6. Dade, H.A. Anatomy and Dissection Honeybee; Bee Research Association Publications: London, UK, Greco, M.K. Imaging techniques for improved bee management. ALP Sci. 2010, 534, Rademacher, E.; Fahlberg, A.; Raddatz, M.; Schneider, S.; Voigt, K. Galenics: Studies toxicity and distribution sugar substitutes on Apis mellifera. Apidologie 2013, 44, [CrossRef] 9. Kanao, T.; Okamoto, T.; Miyachi, Y.; Nohara, N. Pntal exposure to low dose X-rays in Drosophila melanogaster includes early emergence in fspring, which can be modulated by transplantation polar cytoplasm. Mutat. Res. 2003, 527, 1 6. [CrossRef] 10. Rademacher, E.; Harz, M. Oxalic acid: Toxicology on Apis mellifera. Apidologie 2009, 40, Aliano, N.P.; Ellis, M.D.; Siegfried, B.D. Acute contact toxicity oxalic acid to Varroa destructor (Acari: Varroidae) and ir Apis mellifera (Hymenoptera: Apidae) hosts in laboratory bioassays. J. Econ. Entomol. 2006, 99, [CrossRef] [PubMed] 12. Carrasco-Letelier, L.; Mendoza, Y.; Ramallo, G. Acute contact toxicity oxalic acid on honeybees in southwestern zone Uruguay. Chil. J. Agric. Res. 2012, 72, [CrossRef] 13. Desneux, N.; Decourtye, A.; Delpuch, J.M. The sublethal effects pesticides on beneficial arthropods. Annu. Rev. Entomol. 2007, 52, [CrossRef] [PubMed] 14. Rouibi, A.; Bouchema, W.; Loucif-Ayad, W.; Achou, M.; Soltani, N. Risks assessment two acaricides (fluvalinate ad oalic acid) in Apis mellifera intermissa (Hymenoptera, Apidae): Acethylcholinestease and glutathione S-trasferase activities. J. Entomol. Zool. Stud. 2016, 4, Brosgaard, C.J.; Jensen, S.E.; Hansen, C.W.; Hansen, H. Spring treatment oxalic acid in honeybee colonies as Varroa control. DIAS Rep. Hortic. 1999, 6, Gregorc, A.; Pogaènik, A.; Bowen, I.D. Cell death in honeybee (Apis mellifera) larvae treated oxalic acid. Apidologie 2004, 35, [CrossRef] 17. Hatjina, F.; Haristos, L. Indirect effects oxalic acid administration by trickling method on bee brood. J. Apicult. Res. 2005, 44, [CrossRef] 18. Moosbeckher, R.; (AGES, Vienna, Austria); Rademacher, E.; (Freie Universität Berlin, Berlin, Germany). Personal communication, Higes, M.; Meana, A.; Suz, M.; Llorente, J. Negative long-term effects on bee colonies treated oxalic acid against Varroa jacobsoni Oud. Apidologie 1999, 30, [CrossRef] 20. Martin-Hernandez, R.; Higes, M.; Perez, J.L.; Nozal, M.J.; Gomez, L.; Meana, A. Short term negative effect oxalic acid in Apis mellifera iberiensis. Span. J. Agric. Res. 2007, 5, [CrossRef] 21. Nozal, M.J.; Bernal, J.; Gomez, L.; Higes, M.; Meana, A. Determination oxalic acid and or organic acids in honey and in some anatomic structures bees. Apidologie 2003, 34, [CrossRef] 22. Nanetti, A.; Ghini, S.; Gattavecchia, E.; Bartolomei, P.; Marcazzan, G.L.; Massi, S. Pharmacodynamics Oxalic Acid and Treatment Residues in Honey; European Group for Integrated Varroa Control (CA 3686): Rauischholzhausen, Germany, Raddatz, M.; Rademacher, E. Sublethal effects oxalic acid on Apis mellifera L. (Hymenoptera: Apidae): Pharmacodynamics. Apidologie 2010, 41, Schneider, S.; Eisenhardt, D.; Rademacher, E. Sublethal effects oxalic acid on Apis mellifera L. (Hymenoptera Apidae): Changes in behavior and longevity. Apidologie 2012, 43, [CrossRef]

17 Insects 2017, 8, O Neil, M.J. The Merck Index: An Encyclopedia Chemicals, Drugs and Biologicals, 13th ed.; Whitehouse Station: New York, NY, USA, Charriere, J.D.; Imdorf, A. Oxalic acid treatment by trickling against Varroa destructor: Recommendations for use in Central Europe and under temperate climate conditions. Bee World 2002, 83, [CrossRef] 27. Toomemaa, K.; Martin, A.; Williams, I. The effect different concentrations oxalic acid in aquaeous and sucrose solution on Varroa mites and honey bees. Apidologie 2010, 41, [CrossRef] 28. Aliano, N.P.; Ellis, M.D. Bee-to-bee contact drives oxalic acid distribution in honey bee colonies. Apidologie 2008, 39, [CrossRef] 29. Rademacher, E.; Harz, M.; Schneider, S. The development HopGuard as a winter treatment against Varroa destructor in colonies Apis mellifera. Apidologie 2015, 46, [CrossRef] 2017 by authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under terms and conditions Creative Commons Attribution (CC BY) license (